Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(651)

Issue 8783001: Fix the sync sign in dialog issues for app notifications: (Closed)

Created:
9 years ago by Munjal (Google)
Modified:
9 years ago
CC:
chromium-reviews, jstritar+watch_chromium.org, ncarter (slow), akalin, Raghu Simha, Erik does not do reviews, mihaip+watch_chromium.org, Aaron Boodman, tim (not reviewing)
Visibility:
Public.

Description

Fix the sync sign in dialog issues for app notifications: - Call PSS::ShowLoginDialog directly from app notification code so that the dialog comes up even if the user is logged in. - Small fix to PSS to avoid infinite spinner in cases where the sync backend does not really change anything (ahd hence does not call back PSS). Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=112840

Patch Set 1 #

Total comments: 4

Patch Set 2 : '' #

Total comments: 2
Unified diffs Side-by-side diffs Delta from patch set Stats (+17 lines, -1 line) Patch
M chrome/browser/extensions/app_notify_channel_ui.cc View 1 chunk +3 lines, -1 line 0 comments Download
M chrome/browser/sync/profile_sync_service.cc View 1 1 chunk +14 lines, -0 lines 2 comments Download

Messages

Total messages: 9 (0 generated)
Munjal (Google)
9 years ago (2011-12-02 21:33:18 UTC) #1
Munjal (Google)
9 years ago (2011-12-02 21:37:02 UTC) #2
asargent_no_longer_on_chrome
lgtm
9 years ago (2011-12-02 21:41:37 UTC) #3
lipalani1
LGTM from my side. Here are some issues that you want to close before check ...
9 years ago (2011-12-02 21:46:39 UTC) #4
Andrew T Wilson (Slow)
http://codereview.chromium.org/8783001/diff/1/chrome/browser/extensions/app_notify_channel_ui.cc File chrome/browser/extensions/app_notify_channel_ui.cc (right): http://codereview.chromium.org/8783001/diff/1/chrome/browser/extensions/app_notify_channel_ui.cc#newcode119 chrome/browser/extensions/app_notify_channel_ui.cc:119: service->ShowLoginDialog(); Do we need to worry about non-desktop platforms? ...
9 years ago (2011-12-02 21:55:41 UTC) #5
Munjal (Google)
http://codereview.chromium.org/8783001/diff/1/chrome/browser/extensions/app_notify_channel_ui.cc File chrome/browser/extensions/app_notify_channel_ui.cc (right): http://codereview.chromium.org/8783001/diff/1/chrome/browser/extensions/app_notify_channel_ui.cc#newcode119 chrome/browser/extensions/app_notify_channel_ui.cc:119: service->ShowLoginDialog(); On 2011/12/02 21:55:41, Andrew T Wilson wrote: > ...
9 years ago (2011-12-03 01:25:33 UTC) #6
Andrew T Wilson (Slow)
lgtm
9 years ago (2011-12-03 01:39:00 UTC) #7
tim (not reviewing)
http://codereview.chromium.org/8783001/diff/7001/chrome/browser/sync/profile_sync_service.cc File chrome/browser/sync/profile_sync_service.cc (right): http://codereview.chromium.org/8783001/diff/7001/chrome/browser/sync/profile_sync_service.cc#newcode1491 chrome/browser/sync/profile_sync_service.cc:1491: OnAuthError(); Why did we do this this way? From ...
9 years ago (2011-12-06 23:39:31 UTC) #8
Munjal (Google)
9 years ago (2011-12-07 00:05:14 UTC) #9
http://codereview.chromium.org/8783001/diff/7001/chrome/browser/sync/profile_...
File chrome/browser/sync/profile_sync_service.cc (right):

http://codereview.chromium.org/8783001/diff/7001/chrome/browser/sync/profile_...
chrome/browser/sync/profile_sync_service.cc:1491: OnAuthError();
On 2011/12/06 23:39:31, timsteele wrote:
> Why did we do this this way?  From my email, I suggested "...making it so you
> can refresh credentials in the SyncManager, and be notified asynchronously of
> that finishing right up to the top of the stack that originated the
> UpdateCredentials call in the first place (in ProfileSyncService). "
> 
> This seems like more of a hack, so I'm curious what obstacles there were with
> the other route (changing it in sync manager).

The main reason is that this seemed the safest (lowest risk) thing to do in
various discussions with Lingesh and Drew. In fact, Lingesh and Drew
independently mentioned the same approach.

FWIW, our analyses of the code concluded that actually PSS does not need to wait
for a callback from SBH::UpdateCredentials since that only ends up calling
gettime, which is not an authenticated call anyway. Lingesh and Kevin and I
concluded that.

So, if we are right, the right change to do would be to fix that PSS <-> SBH
communication when a user is authenticated.

Powered by Google App Engine
This is Rietveld 408576698